Previous longitudinal studies examined the impact of certain neuropsychological functions such as working memory and response inhibitions on initiation of alcohol use behaviors (Aytaclar et al., 1999
; Iacono et al., 1999
; Nigg et al., 2006
; Thush & Wiers, 2008
). Our study extended these studies and, for the first time, investigated one specific neuropsychological function, affective decision making, in the development of alcohol use behaviors. Moreover, most of the previous studies examined brain function in at-risk American adolescents. Our study applied laboratory-based neuropsychological assessments to the normal adolescent population in China, extending the generalizability of the relationship between affective decision making and substance use across cultures.
Previously, we found a synchronous association between IGT performance and IGT (Johnson et al., 2008
), suggesting but not establishing a causal relationship between affective decision capacity and binge drinking. The study reported here assessed the prospective nature of this association. As predicted by our hypothesis, those adolescents who progressed to the binge drinking or exhibited consistent binge drinking behaviors performed more poorly on the IGT relative to never and occasional drinkers. Moreover, affective decision making significantly and negatively predicted the number of drinks and drinking problems prospectively independent of previous drinking behaviors, working memory capacity, and impulsivity. The results support our hypothesis that affective decision making can serve as a marker predictive of progression of drinking behaviors among adolescents and indicate that dysfunction in affective decision-making capacity renders youth more vulnerable to compulsive and problem drinking behavior.
A substantial proportion of new (83.3%) and consistent (90.9%) binge drinkers performed poorly on the IGT, which demonstrates that the group effect of the poor IGT performance is not due to a few extreme cases. Approximately 53% of consistent never drinkers and 67% of occasional drinkers also performed poorly on the IGT. This is consistent with several other reports of poor IGT performance in adolescents (Crone & van der Molen, 2004
; Hooper et al., 2004
; Overman et al., 2004
). Interestingly, we found that the binge drinkers (new and consistent) even performed significantly worse on the IGT than a subgroup of occasional drinkers who performed poorly (). These results, together with others from our regression analysis, synergistically demonstrate that although all adolescents might still be undergoing developmental changes in the prefrontal region (i.e., having a “premature” prefrontal cortex) (Giedd, 2004
; Gogtay et al., 2004
), by and large be mostly susceptible to making suboptimal or “risky” decisions compared to adults, there is still individual variability among the adolescents at similar age. And those who are at a higher risk for making bad decisions may potentially and eventually translate into real-life risky behaviors, such as alcoholism and drug dependence.
Two previous cross-sectional studies have reported the relationship between affective decision making and drinking behaviors in adolescents and young adults. Overman et al. (2004)
found that polysubstance use was negatively correlated with performance on the IGT analogous task among adolescents. Recently, Goudriaan et al. (2007)
examined the relations between decision making measured by the IGT and binge drinking patterns in young adults over a 2-year time period. They found that stable high binge drinkers performed poorly on the IGT compared to the consistent low binge drinkers (Goudriaan et al., 2007
). Our results are consistent with these previous studies. Moreover, via
the longitudinal study design, our study further supports the notion that a poor capacity for affective decision making is a predisposing factor for alcohol use and abuse.
The prevalence of drinking behaviors in this Chinese sample is similar to that of one large-scale study in China (Xing et al., 2006
). The average number of drinking problems reported by the consistent binge drinkers in our study was only 5 (from a list of 23) at Time 2, suggesting that our study may capture an early stage in progression across abuse trajectories. Interestingly, in our study, the new and consistent binge drinkers both performed poorly on the IGT and scored higher in urgency compared to never and occasional drinkers. Moreover, we found that the IGT net score negatively correlated with urgency. One important difference between urgency and other domains of impulsive behavior is its reliance on emotional factors, and it addresses the role of affect in general and negative affect in particular in the impulsivity behavior (Anestis et al., 2007
). Therefore, the results from the neuropsychological task and psychological measures together indicate that adolescents with better decision making may learn to control the drive from their emotions especially the negative emotions better than those with poor decision-making skills. Moreover, those adolescents who are likely to respond imprudently to their negative emotions may increase their propensity to engage in substance use behavior.
Our results suggest that affective decision making and emotion regulation may share the common neural pathways. This notion is supported by previous research suggesting that neural systems subserving affective decision making overlap with those that support emotional intelligence (EI) (Bar-On et al., 2003
). In addition, a recent functional magnetic resonance imaging study confirmed that adolescents with high EI responded to emotionally provocative stimuli with less total but more focal brain activation in the brain regions that are critical for affective decision making than did their peers with poorly developed EI (Killgore & Yurgelun-Todd, 2007
). Taken together, these studies suggest that adolescents with well-developed affective decision making may possess more efficient and effective functioning neural circuitry than do those with poorly developed affective decision making. They might drink recreationally as did the occasional drinkers and quitters in our sample, but their better affective decision-making capacity protects them from abusive drinking. However, since behavioral tests are not sufficient to identify the brain’s involvement with specificity, further research using imaging methods should address differences between brain circuitry underlying affective decision making in those adolescents who progress toward greater substance abuse compared to those who do not.
Since prior studies have shown that working memory is highly related to general cognitive functions such as reading, mathematics, and reasoning (Colom et al., 2004
; Jarrold & Towse, 2006
), it is not surprising to see that working memory was highly associated with school academic performance in this study. However, neither working memory nor academic performance correlated to any subscales of impulsivity. These results are consistent with prior observations that working memory tasks activate the DLPC-related neural circuitry associated with “cold” cognitive systems (Funahashi, 2006
). Moreover, we found that neither working memory nor academic performance differed in adolescents with various drinking trajectories. Our results suggest that affective decision-making impairment in binge drinkers may be independent of the type of general cognitive intelligence impairment, most likely revealed by tests of working memory capacity (i.e., fluid intelligence).
A limitation of the current study is reliance on self-reports of binge drinking, raising the question of whether a culturally driven self-presentation bias may have influenced the results. A recent study found a binge drinking prevalence rate of 13.2% in the 32,425 senior high school students across 18 provinces of China (Xing et al., 2006
) comparable to those found in our study, 11% at 10th grade and 12.7% at 11th grade. Alcohol use in Western students differs significantly across racial/ethnic groups (Wallace et al., 2002
; Wonga et al., 2004
). Although the rate of alcohol use in our study is lower than the overall rate of U.S. sample, it is comparable to that of Asian students reported in the United States both national and regional studies (Wallace et al., 2002
; Wonga et al., 2004
). The rate of alcohol use is also similar to that of high school students in Canada (Feldman et al., 1999
). Moreover, empirical studies have also shown that the self-reported data are, by and large, valid across racial, ethnic, and cultural groups (Johnston et al., 1994
; Wallace & Bachman, 1993
). Therefore, it is unlikely that there is cultural bias to underreport the socially undesirable behavior. Furthermore, students with low IGT scores may have been more prone to report binge drinking compared with those with high IGT scores, regardless of actual drinking behavior. However, if there is a social desirability self-presentation bias in favoring those with high IGT scores, this bias should be reflected in other socially undesirable behavior as well, such as their poor school academic performance. However, students with low IGT scores did not self-report poorer academic performance than those with high IGT scores ( and ). Taken together, it is unlikely that our findings were greatly affected by culturally or neurocognitively driven self-presentation bias.
There are other limitations in this study. First, the sample size of adolescents with substance use is relatively small, thus limiting the power to detect gender effects in the current study. And the prevalence of drinkers did not differ significantly from Time 1 to Time 2 (). However, the prevalence of substance use in our sample was very similar to that of other large-scale population studies of students in China (Xing et al., 2006
). Second, our study included students from both major types of Chinese high schools, which suggests that the findings are widely generalizable to Chinese youth. Future studies are needed to establish reliability to other cultural/environmental settings. Third, the longitudinal design of the present research covered only 1 year, so we do not know if the binge drinkers in our study would eventually progress to even more problematic drinking behavior. However, previous research showed that substance use behaviors among adolescents significantly predict their subsequent substance use in adulthood (Johnson et al., 2000
). Nevertheless, to our knowledge, this study is among the first to identify neuropsychological risk factors for substance use behaviors among a general adolescent population. Moreover, the longitudinal design provides evidence directly supporting the notion that the specific neuropsychological function, affective decision making, could serve as a unique marker to predict escalated alcohol use among adolescents.